Leaf powerwall has some issues

Nappe

New member
Joined
Nov 14, 2018
Messages
7
Hey folks. I need your input in regards to some troubleshooting on my batteries. My setup is as follows:
I have 2 stacks of Leaf batteries that are connected in parallel. plus Pole from one battery to plus pole on the other battery and from there up to inverter. Same with the negative pole. The minus pole runs through a shunt and they both run through a shunt trip.
My Batrium Watchmon draws 48v from one stack and the shunt trip gets its trip impulse at 24v from the same stack. See diagram.
The cells are connected to busbars of copper pipes from the recycling site which I have cut into lengths and flattend with a screw clamp.
The whole setup runs fine from 3.7v and up and all cells are within 0.01-0.03 volts.
When I discharge the batteries and they reach about 3.7v, cells no. 1-2 and 15-16 run off and lose much more voltage than the others (see picture from watcmon) I thought at first that cell 15-16 was bad and changed those with some new ones but they behave the same way. All cells are measured out with a capacity of about 40Ah + - 2Ah.
It is the cells that are at the top of each stack and those that are closest to the positive pole that protrude. The cable length from plus pole to inverter is somewhat shorter than that from minus pole to inverter as minus is at the bottom of the stack and plus at the top
I also shortened the cable from + Pole to + Pole down so it is roughly the same length as that on the negative poles.
Does anyone have any suggestions as to why the batteries behave this way?
Hope you can decipher my chart.
 

Attachments

  • 32279FD2-36FB-4691-B47D-CADA1CB8F355.jpeg
    32279FD2-36FB-4691-B47D-CADA1CB8F355.jpeg
    1.8 MB · Views: 57
  • A188332E-F002-4888-BE12-65B679D74ADE.jpeg
    A188332E-F002-4888-BE12-65B679D74ADE.jpeg
    4 MB · Views: 58
  • BE454DA8-BF65-4FD8-B245-39528215572A.jpeg
    BE454DA8-BF65-4FD8-B245-39528215572A.jpeg
    2.4 MB · Views: 56

Roland W

Active member
Joined
Oct 9, 2017
Messages
106
Energy of just individual cells which are in series with others cannot escape into thin air if it is present equally in first place if you say. It can only be transferred to other places (via balance leads or other weaker cells in parallel). Or many times you can have voltage issues, if the connections are not tightened enough (Busbars or balance wires) and resistances are different. You can feel that by holding your finger (or have a thermal cam) on your cell terminals and bus bars and look out for unusually high temperatures compared to other spots.
 

OffGridInTheCity

Well-known member
Joined
Dec 15, 2018
Messages
1,530
Interesting. In the pic, I see 3.46v - 3.65v. That's a 190mv difference as you drop toward 3.6v which is pretty big but not 'fatal' if you are OK with 50v low. I run 49.5v low myself.

You mention switching cells out and same result? In my experience an extra foot or 2 of battery cabling will not cause this kind of behavior. And I don't see anything strange in your wiring diagram. The 48v for Batrium is coming from the battery bank (not just one stack) and the 24v tap is only active for a second when the shunt trips - so I wouldn't expect it to be a constant drain or anything like that.

**Perhaps you have a sharp discharge curve knee... in combination with some of your cells not as capable as the others... so things are tight until you hit 3.6v (or so) and then some of the cells drop very sharply.

I make the above comment because.....
In my own case, I have 84 packs of 260ah each of 18650s. They were built carefully to have the same 260ah +/- 2ah each. However, once they were in operation I found that 10% (8 or 9 packs) has to be 'bolstered' with additional 5-10% capacity to avoid rapid voltage drop at the low end - similar to what you're seeing. What's a bit of a mystery is that these 10 packs were not 'bad' in the sense of self-discharge - they just sagged a bit more than the other 70 packs and once they were tweaked - they were OK.

------------------------------------
Moving forward - just brainstorming....
1) IF you can operate with 'tight balance' from 4.X -> 3.7v and then 190mv divergence at 50.0v cut-off with 3.46v as you're lowest, AND its stays that way cycle after cycle... (e.g. no self-discharge going on) then operationally there's really not much of a worry.

2) Looks like you only have 3 that are really low and perhaps its a longmon issue? Maybe move (swap) some longmons around and see if things stay the same.

3) Perhaps run with 1 stack (instead of 2) to verify that the errant cells continue to misbehave - and then replace those with better cells from the inactive stack and see if things settle out?

4) Add 5 or 10 18650s in parallel to a 'lagging cell' / rebalance and see if boosting individual lagging cell capacity by 5-10% keeps it from sagging? *I just tape / solder a few 18650 cells together and attach to main battery cell with alligator clips - no need to rebuild the whole battery just to tweak capacity a bit to experiment :)


I'll be interested to see other comments and follow this as you research it.





.
 
Last edited:

ajw22

Well-known member
Joined
Nov 16, 2018
Messages
584
Perhaps it's a temperature issue? The cells on the top of the stack will be cooler, and will have a slightly higher IR, leading to a higher voltage drop...? Whereas all other cells are cozily sandwiched between their warm neighbors, and thus will perform better.

Probably not related to the OP issue, but you should reattach the main negative cable to the other end. Should lead to more balanced loading of all cells.
Not a big issue if you have really beefy cables and/or operate at low currents.
1618797786379.png


Perhaps pictures of the actual setup might help?
 

J_Mack58

New member
Joined
Apr 7, 2021
Messages
21
Hey folks. I need your input in regards to some troubleshooting on my batteries. My setup is as follows:
I have 2 stacks of Leaf batteries that are connected in parallel. plus Pole from one battery to plus pole on the other battery and from there up to inverter. Same with the negative pole. The minus pole runs through a shunt and they both run through a shunt trip.
My Batrium Watchmon draws 48v from one stack and the shunt trip gets its trip impulse at 24v from the same stack. See diagram.
The cells are connected to busbars of copper pipes from the recycling site which I have cut into lengths and flattend with a screw clamp.
The whole setup runs fine from 3.7v and up and all cells are within 0.01-0.03 volts.
When I discharge the batteries and they reach about 3.7v, cells no. 1-2 and 15-16 run off and lose much more voltage than the others (see picture from watcmon) I thought at first that cell 15-16 was bad and changed those with some new ones but they behave the same way. All cells are measured out with a capacity of about 40Ah + - 2Ah.
It is the cells that are at the top of each stack and those that are closest to the positive pole that protrude. The cable length from plus pole to inverter is somewhat shorter than that from minus pole to inverter as minus is at the bottom of the stack and plus at the top
I also shortened the cable from + Pole to + Pole down so it is roughly the same length as that on the negative poles.
Does anyone have any suggestions as to why the batteries behave this way?
Hope you can decipher my chart.
This how I see your set up (Autocad). In my much smaller Leaf battery the last positive cell going to my inverter is 3.39v all other cells are 3.44v as the cells charge up the low cell catch up. I been chasing electrons since the days of the Simpson analog meter, now Fluke is King. I’m thought about it and came up with, “That’s just the way it is”. 🥸
 

Attachments

  • 59DA9614-7974-4E71-B620-71009278E2FD.jpeg
    59DA9614-7974-4E71-B620-71009278E2FD.jpeg
    2.2 MB · Views: 25
  • image.jpg
    image.jpg
    1.7 MB · Views: 20

Nappe

New member
Joined
Nov 14, 2018
Messages
7
This how I see your set up (Autocad). In my much smaller Leaf battery the last positive cell going to my inverter is 3.39v all other cells are 3.44v as the cells charge up the low cell catch up. I been chasing electrons since the days of the Simpson analog meter, now Fluke is King. I’m thought about it and came up with, “That’s just the way it is”. 🥸
It’s the excact same thing - celle catch up when i recharge. It just seems to be precious watthours wasted on Those fed celle Tinning Los... Well perhaps not. Hos is your charger setup regarding cutoff an Max?
 

Nappe

New member
Joined
Nov 14, 2018
Messages
7
Here is a picture of my setup. Small wires need tidying:)
Back up at 3,7v they catch up just fine.
 

Attachments

  • 85CB9DC7-7D4E-4BCA-80CE-1EFF3206DD7A.jpeg
    85CB9DC7-7D4E-4BCA-80CE-1EFF3206DD7A.jpeg
    1.2 MB · Views: 42
  • A554B29D-16DC-4BAC-8CAE-8059EF3F793E.jpeg
    A554B29D-16DC-4BAC-8CAE-8059EF3F793E.jpeg
    4.2 MB · Views: 39

slimf

Member
Joined
Sep 26, 2017
Messages
148
Ok, what a mess. :p

Currently, you have TWO, 2P 14S stacks. Start by rebuilding this into ONE, 4P 14S stack. Locking as many in parallel as possible will help reduce cell drift.

I have my Batrium 'Cell Low cut out' set to 3.4v and 'Cell low resume' at 3.7 - However, I find that as the cells go below 3.7 some drop faster than others and effectively the discharging stops while most of the batteries are at 3.7 or above. So in essence, the same issue as you.

When you rebuild your stack into 4P, do a BOTTOM balance. Discharge all down to say 3.4 and then connect them ALL together - let them balance out.. Then disconnect and make into your 4P stack. This should help. I haven't been bothered as I have been working on getting mine to 21P (~40-45Kwh)

Attached is what mine looked like when I had 13P.

Note: I use a WM5 - balance wires on bottom cells in each of the 7 stacks.
 

Attachments

  • IMG_7377.jpg
    IMG_7377.jpg
    2.1 MB · Views: 32
Last edited:

Redpacket

Well-known member
Joined
Feb 28, 2018
Messages
1,265
+1 to ajw22's wiring change
Since the issue happens under load, and only to cells at the physical end of the packs, it might be a physical wiring arrangement issue.
Eg where you have the cell mons attached due to the resistance of the links? Yes the links are copper but they still have definite resistance?
If the longmon measuring points are different to the "mid-pack" cells, they might report different voltages?
Maybe use a multimeter on 200mV range to check for voltage drop across those links when you have a load on the system.

Also, where you have put the longmons ie against metal and in contact with the battery looks dangerous to me. A) if the black paint scrapes they will short circuit to the battery case. B) they will heat the cells in one spot, creating a chemistry temperature difference inside that cell. You want cells to be even temps inside not hotter on one edge. C) the longmons need open air to cool.

Leaf build issues.jpg
 
Last edited:

J_Mack58

New member
Joined
Apr 7, 2021
Messages
21
Ok, what a mess. :p

Currently, you have TWO, 2P 14S stacks. Start by rebuilding this into ONE, 4P 14S stack. Locking as many in parallel as possible will help reduce cell drift.

I have my Batrium 'Cell Low cut out' set to 3.4v and 'Cell low resume' at 3.7 - However, I find that as the cells go below 3.7 some drop faster than others and effectively the discharging stops while most of the batteries are at 3.7 or above. So in essence, the same issue as you.

When you rebuild your stack into 4P, do a BOTTOM balance. Discharge all down to say 3.4 and then connect them ALL together - let them balance out.. Then disconnect and make into your 4P stack. This should help. I haven't been bothered as I have been working on getting mine to 21P (~40-45Kwh)

Attached is what mine looked like when I had 13P.

Note: I use a WM5 - balance wires on bottom cells in each of the 7 stacks.
Real nice looking set up there. I just slapped myself, after all the experimenting and prototyping
Ok, what a mess. :p

Currently, you have TWO, 2P 14S stacks. Start by rebuilding this into ONE, 4P 14S stack. Locking as many in parallel as possible will help reduce cell drift.

I have my Batrium 'Cell Low cut out' set to 3.4v and 'Cell low resume' at 3.7 - However, I find that as the cells go below 3.7 some drop faster than others and effectively the discharging stops while most of the batteries are at 3.7 or above. So in essence, the same issue as you.

When you rebuild your stack into 4P, do a BOTTOM balance. Discharge all down to say 3.4 and then connect them ALL together - let them balance out.. Then disconnect and make into your 4P stack. This should help. I haven't been bothered as I have been working on getting mine to 21P (~40-45Kwh)

Attached is what mine looked like when I had 13P.

Note: I use a WM5 - balance wires on bottom cells in each of the 7 stacks.

Ok, what a mess. :p

Currently, you have TWO, 2P 14S stacks. Start by rebuilding this into ONE, 4P 14S stack. Locking as many in parallel as possible will help reduce cell drift.

I have my Batrium 'Cell Low cut out' set to 3.4v and 'Cell low resume' at 3.7 - However, I find that as the cells go below 3.7 some drop faster than others and effectively the discharging stops while most of the batteries are at 3.7 or above. So in essence, the same issue as you.

When you rebuild your stack into 4P, do a BOTTOM balance. Discharge all down to say 3.4 and then connect them ALL together - let them balance out.. Then disconnect and make into your 4P stack. This should help. I haven't been bothered as I have been working on getting mine to 21P (~40-45Kwh)

Attached is what mine looked like when I had 13P.

Note: I use a WM5 - balance wires on bottom cells in each of the 7 stacks.
Real nice looking set up there. I just slapped myself twice... after all the experimenting and carving out a spot to put my batteries I need to make it look nice like yours. Got to take the time to do it. Got to take the time to do it. Got to take the time to do it.
 

ajw22

Well-known member
Joined
Nov 16, 2018
Messages
584
The whole setup runs fine from 3.7v and up and all cells are within 0.01-0.03 volts.
When I discharge the batteries and they reach about 3.7v, cells no. 1-2 and 15-16 run off and lose much more voltage than the others [...] All cells are measured out with a capacity of about 40Ah + - 2Ah.

I suspect your swapping out test was not done correctly. Are you sure the new modules had same/higher capacity as all the others, and that all cells were _fully_ top balanced?

I think what's you're seeing is just normal with the cells you have. The discharge curve(*1) of Nissan Leaf cells have a very pronounced "knee" and start to drop like a rock at around 3.65v. With cell capacities ranging from 38Ah ~ 42Ah (10% diff!), voltages are going to diverge very quickly and dramatically as the lowest capacity cells drain past that knee.
(*1) https://insideevs.com/news/323768/dissecting-the-nissan-leaf-battery-w-video/

Two solutions:
1.) Stop discharging at 3.70 volts. There's not much energy below that anyway, and your cells are going to last much longer.
2.) You have 2 Leaf modules in parallel. Figure out _exactly_ how much capacity each individual module has, then re-match / rearrange the pairs in such a way that the SUM of each 2 module pair is the same. Something like this would be really handy for figuring out the capacity:
 
Last edited:
Top